node_runner

Just a couple of random thoughts as a result of reading this thread: -Without having a good view of why an AI is behaving a certain way, it is difficult to judge whether the AI is being stupid or smart. If there is some bit of information which is known to the AI, but is not known to you as a human observer, then sometimes AI behavior can appear to be dumb when in fact there is good reason for the behavior. For example, watching a BLUFOR squad retreat from a firefight that they are clearly winning may seem dumb if you don't know that the squad is running very low on ammo. A good exercise is to imagine yourself as an alien observing human behavior on Earth from your spaceship, without knowing anything about human society. -The only valid way to judge the quality of AI is by how well it achieves the goals that the developer intends to achieve, not by how effective it is at defeating you. -Here is a cool site to learn about gaming AI and some of the techniques and challenges: http://www.aiwisdom.com/ -While I agree that ArmA2 is a much more ambitious game as far as AI requirements go than most other games, I wouldn't put it in my top 5 list of most impressive AI. Consider the following games as a comparison of ambitious AI that pull off their intended goals with impressive execution: Elder Scrolls 4: Oblivion, FEAR, The Sims 2/3, Fallout 3, Thief: The Dark Project, Left 4 Dead, GTA IV. They aren't apples to apples comparisons, but when I think of AI that made me feel like I was in a living world, I think these are better examples. -I think the AI in ArmA2 would greatly benefit from two additional AI "systems". I've been thinking about trying to do this myself in a mod, but I don't know how feasible it is or whether or not I'd be up to the task. The two features would be: 1) Overall Strategic Planning w/ Strategic Waypoints The AI does not navigate the terrain with any military strategy. If you set a waypoint down, they simply route to that point in a straight line, working around obstacles as they are encountered with basic pathfinding routines. It would be ideal to use the mission editor and define higher level objectives, but allow the AI to plan on its own how to go about navigating the terrain to complete that objective. For example, giving an AI squad a "Seek & Destroy" objective instead of a waypoint, the AI would plan a more strategic route across the terrain towards that objective and plan that route keeping things like the following in mind: -Avoid skyline routes -Scout unknown terrain *before* crossing it -Determine a good attack vector for the objective, keeping in mind cover, concealment, height advantage, and range. -Scout the target area in stealth mode and designate important targets before engaging. -Modify formations and behavior modes (danger, safe, stealth, etc.) dynamically as appropriate. 2) When commanding your AI, it would be nice to specify the "firmness" of your commands. Just having a "firm" or "soft" version of most commands would be best. Firm, meaning "follow my orders literally and do not deviate as you see fit", soft meaning "follow my orders as a guide, but modify as you think you need to". I would imagine just using a modifier like the SHIFT key to each command in the command menu. By default, each command would be "soft", but if I held the SHIFT key while giving the command, it would be "firm". So, for an example situation, sometimes when I tell an AI to engage a target, I want to specify one of two different styles: Soft: Engage the target as you see fit, breaking formation if need be, navigating to different terrain as you see fit, and planning on how to take down the target on your own. If the enemy moves behind cover, figure out a way to flank him. Firm: Start shooting at the target but don't break formation or move around on your own. If the enemy finds cover, stay put.

I'm also having the problem where the AI will fire on distant enemies while in HOLD FIRE mode.

It seems that A2 isn't getting any performance benefits from crossfire when using Radeon 5870 cards. I have 3 Radeon 5870 cards in a crossfire configuration. Every other game I own has an incremental performance boost from 40%-80% each time I activate another one of the cards. But with A2, I get no performance benefits at all over using a single 5870. I have tried four versions of the Catalyst drivers so far with no luck: 9.10, 9.11, 9.12, and 9.12-hotfix. I'm using the 1.05 version of A2 and running the "Benchmark 1" scenario mission to gauge performance. Here is an example of crossfire scaling on my system using Dirt2 as a benchmark: Crossfire disabled (using a single card): 55fps avg. Crossfire enabled using x2 cards: 90fps avg. Crossfire using x3 cards: 115fps avg. When I run the "benchmark1" scenario in A2: Crossfire disabled (using a single card): 52fps avg. Crossfire enabled using x2 cards: 51fps avg. Crossfire using x3 cards: 51fps avg. I realize this may have more to do with ATI/AMD drivers than anything else and that BIS may not be able to do anything about it. I've also posted this problem on the ATI forums, but I want to post here in case there is anything BIS can do to notify ATI/AMD of the issue so that it can be addressed in a future driver update. I'm not sure what else I can do to get the attention of the right people to have this looked at, but any suggestions are appreciated. Has anyone else experienced similar issues with the 5800 series cards using crossfire? Am I the only one having this problem? Every other game and benchmark I have seems to scale properly. A2 is the only problem. Thanks. *Edit: I didn't post this in the existing performance issues thread because on a single card A2 performs fine. I don't think this is actually performance issue but something related to a driver bug with ATI or a bug/addition which needs to be made to A2 so that it works with 5800 series crossfire.

Are they actually coming in as reinforcements or because they were alerted for some reason? When you go to the map and watch the units, does it actually say "Reinforcement" on the unit symbol?

The point is that inside your ../arma2/UserConfig folder, you want to have a "GL4" folder with the settings files in them. All arma2 mods will tend to put settings that the user can edit in side the ../arma2/Userconfig/<modname> folder. If you already have the folder there you can just drag it over and use "copy & replace" in windows and it should be fine. Just make sure you have a ../arma2/userconfig/GL4 folder with stuff inside of it and you will be ok.

It would be cool if you could make access to a radio a requirement for AI to call reinforcements. Maybe you could set up two different ranges for reinforcement radius in the config file. One for long range radios and one for short-wave radios. That way soldiers carrying the standard short-wave can call any reinforcements in a range of maybe 300m, but if you have a radio backpack or if you are in a vehicle, you can call for reinforcements in a much larger range, like the default 50000.

Wonderful. Even a small step in the right direction will have a big payoff. I'm so interested in getting this part of the AI fixed that it is motivating me into learning how to modify the AI. I already have basic C++ skills, so I've started reading up on the "FSM" section of the BI wiki. If you have any other suggestions for good learning resources, it would be appreciated. I'm going to upack the GL4 and ZeusAI pbo's to see what I can learn from those as well. Thanks for all your effort, and let me know if there is anything I can do to help.

It looks like the latest version of GL4 has changed this setting around yet again. In my config file, array 36 seems to be what you want to turn off: //GL4 Enemy A.I. Difficult Level: //======================= //Choose False to disable this feature completely. // True/False, default is True GL4_Global set [36, False]; However, after doing this and making sure that ZeusAI gets loaded *after* GL4, I'm still having problems with AI not firing at each other at very close CQB ranges. This problem seems to go away when I take out GL4 from my mod list and just test ZeusAI along with ACE. EDIT: I've reloaded everything again and now it seems to be behaving properly. Does white space matter in these files? If so, they GL4_Global line was indented which was causing the problem. Also, I've noticed that without a waypoint, the peripheral vision and ability of the AI to sense enemies close to them but not right in front of them is very poor. If you load up TM2 and drop in a BLUFOR without a waypoint, then spawn an OPFOR on a 2 o'clock position from the BLUFOR, they will not detect each other due to very poor peripheral senses. But if you give BLUFOR a "Guard" waypoint, his peripheral perception increases dramatically. He can sense and react an enemy infantry on his 3 or 9 o'clock positions as you would expect a real person to do. A few other things I've noticed about Zeus AI: -AI still rarely uses smoke to withdraw. -Grenadiers rarely fire their GL's when enemy AI is clustered into a group. -Medics blindly rush into the line of fire to heal wounded and get shot themselves. I have yet to see a medic attempt to drag or carry wounded to a safer position. Suggestion: There is fairly active group of people here who are committed to help in any way we can to improve ZeusAI and GL4 (and how they interact with each other). Some ways that we might be able to help you guys continue to improve your wonderful work: -Set up a bug/feature tracker on dev-heaven.net that we could all post bugs and feature requests to. -Come up with a standardized set of SP missions that are good for testing the AI under certain conditions that we all could run and report back the results. Troopmon2 would be useful for this. For example, I use TM2 to spawn an OPFOR and BLUFOR right next to each other and watch how long it takes for them to notice each other and fire on one another. Maybe we could help you guys put together a little mission pack that would test the AI in all kinds of conditions, then we could all run them and report back odd or broken behavior.

Amazing addon. I started playing with this yesterday and I was so impressed I went through all 70 pages of this thread so I can learn more of how it evolved and how to use it better. GL4 has become as essential for me as ACE over the course of a single night. In an earlier post, you said that fixed the issue that was causing BLUFOR and OPFOR to attack civilians when using the default CBA method of loading GL4. I've been using the module method so that I have more control over how it works, and I notice an issue. When I play as a civilian to monitor the AI, they don't attack me, but they are threatened by me. BLUFOR goes into combat mode as soon as they see me, and even pop smoke sometimes to get away from me. But they don't actually fire on me. I also have a request that could turn out to be fairly ambitious but I think it is one of the biggest things holding the AI back from behaving more human. The idea is simply that the AI doesn't prioritize which direction they should be watching very well. They will often take cover and start watching the opposite direction of where they are getting attacked from. I understand that they scan their flanks to try and avoid being flanked from the sides, but they often go prone and just watch in the complete opposite direction of the incoming fire which is a waste. I don't know if the best way to handle this is to have some kind of array to keep track of which areas are likely to be safe, hostile, or "unknown", but it seems like you could do a lot with something like that. The biggest improvement right now would be for the AI to know "I'm taking fire from the north, so I will take cover and try and watch or return fire to the north, and occasionally check NE and NW for additional contacts, but I will completely ignore the South because that is the direction I came from". But if the AI kept an array of surrounding areas which they updated periodically as being either hostile, safe, or unknown, they could also move through the terrain smarter. For example, they could be moving across a flat terrain that they know is "safe" because they already scouted that direction. Then when they start to crest a hill, the area beyond that might be considered "unknown", so they go into stealth mode, crest the hill, scout ahead, and mark the array as either "safe" or "hostile" and then proceed accordingly. Likewise, if you gave an AI group a waypoint to an area that they already know is hostile, they could do a better job of approaching cautiously, and from a direction that gives them some cover, concealment, and/or height advantage. Bonus points if you could get them to go into "hold fire" mode while they get into position, and then open fire once they all are near cover and have a target in sight. I know it sounds ridiculously ambitious, but it only in the sense that it would be a lot of coding. It doesn't seem impossible to me, but I don't know enough yet about modifying the AI behavior. Anyway, take that for what it is worth and thanks so much for GL4!

Correct. I said that I wasn't done yet in my post, but it is enough to get a good idea and do some comparisons. There is more than half the data there. I have yet to do 2.5 of the configurations out of a total of 12.

Benchmark 1 and 2 can be found in the Scenarios section within Single Player (assuming your version of A2 is patched). What do you mean by "not an accurate test"? All of those benchmarks simply place you in the game with a fixed camera. The game engine is rendering the same objects that it would be if you were controlling a player yourself. But I don't even know what you mean by "accurate" benchmark so you'll have to please explain.

Sorry that this is taking me so long. I had some problems where I started to get very inconsistent results and I had to re-do a ton of my benchmarks due to arma2.cfg issues. I'm not completely done yet, but I've finished enough of it to get a good idea of how much the CPU and GPU's are related and how much they affect performance at least on my machine. I'll update the Google Doc as I finish more of the benchmarks, but please bear with me because I have to run each of those benchmarks multiple times to make sure it is consistent and something odd isn't happening. Setup My hardware specs are: -Intel Core i7 Quad Core CPU @2.6Ghz stock -Water cooled cpu, capable of pretty good overclocks. -x3 Radeon 5870's -6GB Triple Channel DDR3 Corsiar memory at 1600mhz. -Windows 7 x64 -Windows installed on a pair of Velocirator HDD's in raid 0 -Arma2 installed on dedicated Intel X25M 80GB SSD -Single 27" monitor with a native res of 1920x1200 Software Info: -Arma2 version 1.05 vanilla (no mods used for benchmarks). -ATI Catalyst drivers version 9.12 (non-hotfix version). -ATI Tray Tools 1.6 Beta (used to monitor FPS and force-disable vsync in Arma2, as well as monitor usage of each GPU) -Built in Windows 7 performance monitor used to monitor CPU, memory, and SSD usage. ArmA2 Configuration: -Hyperthreading disabled in BIOS -Vsync forced off using ATI Tray Tools mentioned above -Catalyst AI set to "Advanced" -Executable started with the following options: "-winxp -maxmem=2047 -cpuCount=4 -nosplash" -Arma2.cfg modified to specify the localvram settings depending on how many GPU's I'm using for my test: mem settings with 1 card: 536870912 mem settings with 2 cards: 1073741824 mem settings with 3 cards 1610612736 Two different sets of video settings used for benchmarks: [Low Quality] Draw Distance: 500 Texture Detail: Low Video Memory: Default (uses size specified with localvram in arma2.cfg) Anisotropic filtering: Disabled Antialiasing: Disabled Terrain detail: Very low Objects Detail: very low Shadow Detail: disabled Posprocess effects: Disabled Resolution: 960x600 [High Quality] Draw Distance: 1600 Resolution: 1920x1200 Texture detail: very high Anisotropic filtering: very high Antialiasing: high Terrain Detail: very high Objects detail: very high Shadow Detail: very high Post Process: very high Benchmarks used: -Benchmark1 (tests zoomed in battle during the day) -Benchmark2 (tests zoomed out battle at night) -Arma II Mark (custom mission designed to perform multiple tests and give a final benchmark index at the end). Testing Methodology Because people seem to have different opinions on how well A2 takes advantage of multiple GPUs, how limited it is by CPU's, etc., I wanted to set up benchmarks to put those claims to the test. I wanted to run all three benchmarks on both High Quality settings and Low Quality settings, testing them on 1 GPU, 2 GPU's, and 3 GPU's. But since a lot of folks have been saying it is a CPU limitation, I also wanted to run all sets of benchmarks before and after overclocking my CPU. Some of the benchmarks have my system running the CPU stock at 2.6Ghz. Another set has them running overclocked at 3.3Ghz. If the CPU really is the bottleneck, then Crossfire *should* show more of a performance advantage with my CPU overclocked. While I ran each test, I also had monitors up to watch the following statistics: -CPU load of each core -GPU load on each video card -Memory usage -%Idle time of the SSD that A2 is installed on My thought was that I should be able to see the GPU's get utilized more once my CPU was overclocked. I ran each benchmark, and make some short notes on what I saw with the different monitoring tools. Results The raw FPS numbers of my benchmarks so far can be viewed on this Google Docs spreadsheet: https://spreadsheets.google.com/ccc?key=0AnNLQZzfYb01dGFPZ0RmR3MtdmF1cGx4MUcxNW9WVGc&hl=en

I tried testing out some of the sniper action last night with a friend in a small test mission. I noticed that while I was using the spotter scope, I couldn't see any of the dust kicked up from his missed shots when they hit the ground around the target, but he said that he could see it. The reverse happened when he was a spotter and I was shooting the rifle. Do those little dust markers not get sync'd across the network in multiplayer? If not, is there some method planned for the spotter to be able to tell where a shot landed so he can direct the shooter on how to adjust? Thanks!

I have a big post coming, with detailed/painstaking benchmarks to put these claims to the test. I've been working on them all day. I'll post them as soon as I'm done.

I am running A2 on a dedicated intel x25M SSD. I know that currently, A2 has some fairly poor memory management and doesn't support x64 addressing, nor will it address over 2GB of memory. That is at the top of my wish list for OA. But I dunno. I have a pretty rockin' system here. Quad-core 3.8Ghz, dedicated SSD, etc. You would think that even though it is a data hog, A2 would still show performance increases for multiple GPU's. Lastly, I don't think the texture data that A2 loads would keep it from utilizing the GPU's. It has to load all that data from the hard drive anyway, no matter how many GPU's I have. I'm not sure I understand why adding another GPU would suddenly cause A2 to start loading twice as much data from the HDD.

That is what I was afraid of. Are you sure about this? I hope this isn't true, because if that is the case, then A2 is unlikely to ever take advantage of multiple GPU's and I'd have to just keep my fingers crossed for OA. Can anyone else confirm this? Anyone with nVidia cards in SLI mode? If you run with the exact same graphical settings with and without SLI/Crossfire, is there a performance difference? If it is just the drivers then I can at least have some hope and continue to pester ATI/AMD. If the game itself just doesn't support it then pestering BIS is going to be my only chance. Considering how much performance this game demands it would seem counter-productive not to support multiple GPU's.

I appreciate your help, I really do. It's nice of you to take the time post suggestions. But my performance is driver-related, and isn't going to be affected that much by vsync. Maybe the difference of like 4-5 fps, but that is about it. As far as testing it on another island instead of benchmark 1, of course results will vary from mission to mission. The performance is heavily related to how many AI there are, how many explosions are happening, where you are on the terrain, how zoomed you are, etc, etc. There is no doubt that my FPS will go up if I take out all of the AI and look around by myself on Utes on a blank map. That isn't really the point. Also, I'm not complaining that the performance is bad. I'm not really looking for tips on how to get a better framerate by modifying the game settings. The game settings themselves don't really matter. Whether I have all of the settings maxed or not, I should still be getting better performance with 3 GPU's than with 1 GPU, but the performance is always the same, independently of what the video options are set to. I have done more than just run benchmark1. This is after weeks of playing the game in various single player and multiplayer session, and there just really isn't any difference with crossfire on or off. The 5000 series has had crossfire problems for some time now. This is a known issue with this series of cards. I'm sure that it is a driver-related problem. It has been cleared up with most games now, but it is still an issue with A2. I wasn't really looking for a solution on these forums, so much as a "Hey, if anyone at BIS is watching, please pester anybody at ATI if you can so they can fix this sooner than later" kind of thing and to see if there are others like me who are trying to run the 5800 series as a crossfire setup with A2. Maybe if there are others we can get enough people complaining about the issue to ATI/AMD and it will be a priority on their list of things to fix/improve in the next driver release.

That isn't the problem. I'm running a quad core i7 @3.8Ghz and the CPU is not stressed during the benchmark. I know how to benchmark my PC. The problem is definitely either driver related or a problem with A2, most likely driver related.

You don't need to define anything if you are using the ambient civilian module on Utes or Chernareus. Just drop the module somewhere on the map and you are good to go. The only reason to change any of the module parameters is if you don't want to use the defaults. I've messed with this a little bit, but if you bring the civilian count too high things get weird and performance drops, so I don't recommend it. But if you do want to tune that stuff, the module is documented on the BIS editing wiki.

As far as I know, you can't use that modules on any other maps than the two BIS ones. I'm still kind of a newb at editing though, so I could be wrong.

Great updates! The ranging information is very helpful, thanks.

I hope I'm not reading too much into the translation when he was talking about things being more destructible. Having a greatly improved destruction-physics system would be on the top of my list if I got to choose what would get upgraded/added. I'm absolutely stoked at the idea of seeing things blow apart and watching tanks and launchers blow holes through walls. Oh man, with major upgrades in a few key areas: -Destruction physics -Flight model -AI That would be it for me. I might have to say goodbye to playing any other game. Sign me up for the $20 monthly subscription to THAT BIS!!!

This is a great guide and a big help for guys like me who have no military experience and don't know much about these weapons. May I suggest that if you have the time or the gumption, including a little blurb about some of the sight markings that you don't cover due to complexity and what they are used for would make for good reading. I understand you don't want to write a whole chapter on leading targets with the horizontal markings on some of the scopes, and that makes sense. But if there is ever a time where you could imagine someone who doesn't know much about these guns wondering "hey, I wonder what that is for or how it is used", maybe just including a few sentences on what that stuff does or maybe a link to a wikipedia article or something. Anyway thanks a lot for putting in your own personal time to do this. It helps guys like me a lot. Another thing that would be cool is if you could point out any significant differences between using the weapon in ACE and in real life. The little tidbit about the Chinese rifles perhaps being more accurate in ACE than they probably are in real life, or just the fact that suppressors heat up a lot in real life and I'm guessing you could damage the weapon or injure yourself if you continued to fire a suppressed rifle when it was getting too hot. Those little bits of information are really fun to read.

I've been wanting to see this since ArmA1 but most people didn't take the idea very seriously when I mentioned it. Nice to see the topic coming up again.

This has been an interesting thread to read. A couple of points: -People should be allowed to do whatever they want in the privacy of their own home provided they aren't violating the rights of anyone else. This includes the depiction of fantasy situations that other people would find immoral. -BUT... -ArmA2 is the intellectual property of BI. If BI decides that they don't want certain mods or edits to be made to their IP, that is totally within their right and very well should be. That includes anything that is posted or discussed on these forums. These forums are private property, and that has to be respected. Private property rights always trump free speech. If BI doesn't want certain things to be depicted with their game, that is their right, whether or not you agree with their reasoning. -I don't think this mod would be a good idea from a *technical* standpoint, because the kinds of scenes that you are describing requires a lot of AI in the map, a lot of custom animation work, and a lot of scripting. All of this adds up to terrible performance due to the way the engine works, especially in multiplayer games. I just don't see how it makes sense to use those precious CPU cycles for something like this, instead of using it for better combat with the AI. This type of thing would fit much better with Fallout 3, which I would recommend you use for this type of mod instead of ArmA2. I know that it is just my opinion that it is better to use the resources for other means, but ArmA2 was built to be focused on infantry and combined arms combat, which is why it makes sense to use it as such. I know the BI developers will post that the engine is so powerful and versatile you can use it for anything, but think about it reasonably. Could you create a mod to turn ArmA2 into a turn based strategy game? Probably. Would it makes sense to do that in ArmA2 instead of Civ4? Absolutely not. Can you pound nails into wood with a screwdriver handle? Sure you can. Does it make sense to do it with a screwdriver instead of a hammer, given the choice? Absolutely not. Separate from what your moral opinions are, I just don't see how ArmA2 is the best tool to use for something like this, rather than other games which seem much better suited (such as Fallout 3).